Our objective is to develop a Gene Expression Database for Mouse Development (GXD). GXD will make gene expression data freely and widely available in formats appropriate for thorough analysis and will thus provide the scientific community with a tool for examining patterns of gene expression during embryonic development, elucidating biologically important molecular networks, and exploring the genetic programs that underlie normal development and disease. GXD will store and integrate data from different expression assays. Primary rather than processed expression information will be stored. In this format, new data can be added readily, novel insights resulting from new data can be represented, and higher order information can be synthesized from the available data. Thus, GXD will capture information about what products are produced by a gene and about where and when these products are expressed. Expression patterns will be described by a controlled anatomical dictionary. Textual descriptions of in situ studies will be complemented with digitized images of original expression data. In the long term, this text/image system will be combined with a 3D atlas of mouse development to enable 3D graphical display and analysis of expression patterns. GXD will be interconnected with other databases containing DNA and protein sequence information, genetic and physical mapping data, and disease descriptions. This will significantly increase the usefulness of the resource. Integration with the Mouse Genome Database will foster a close link to genotypic and phenotypic data, and GXD will build a natural bridge between these data. The specific aims of this proposal are to: l. Design and test database prototypes and to establish the GXD as a community resource. GXD prototypes and interfaces for electronic data submission will be tested extensively and iterative refinements made based on our experience and input from contributing laboratories. 2. Develop query and display tools. Graphical as well as textual querying tools will be developed to cope with the complexity of gene expression data. 3. Populate and maintain the gene expression data in GXD. GXD will be populated with data via electronic submission from laboratories and from annotation of the published literature by the GXD editorial staff. Data accuracy will be reviewed at several levels. User support will be provided for electronic submission, database querying, and to develop user documentation and training.